EPUB Output: Strip <form> tags since ADE runs screaming when it sees one. Fixes #2029 (IHT resetting P505)

src/calibre/ebooks/epub/from_html.py

@@ -197,6 +197,9 @@ def fix_markup(self):
             if not tag.text and not tag.get('src', False):
                 tag.getparent().remove(tag)
                 
+        for tag in self.root.xpath('//form'):
+            tag.getparent().remove(tag)
+                
         if self.opts.linearize_tables:
             for tag in self.root.xpath('//table | //tr | //th | //td'):
                 tag.tag = 'div'

src/calibre/web/feeds/__init__.py

@@ -98,7 +98,7 @@ def populate_from_feed(self, feed, title=None, oldest_article=7,
             if len(self.articles) >= max_articles_per_feed:
                 break
             self.parse_article(item)
-            
+        
         
     def populate_from_preparsed_feed(self, title, articles, oldest_article=7, 
                            max_articles_per_feed=100):
@@ -156,7 +156,6 @@ def parse_article(self, item):
             content = None
         if not link and not content:
             return
-        
         article = Article(id, title, link, description, published, content)
         delta = datetime.utcnow() - article.utctime
         if delta.days*24*3600 + delta.seconds <= 24*3600*self.oldest_article:

src/calibre/web/feeds/news.py

@@ -1012,7 +1012,8 @@ def parse_feeds(self):
                 feed.description = unicode(err)
                 parsed_feeds.append(feed)
                 self.log_exception(msg)
-                
+        
+        
         return parsed_feeds
     
     @classmethod

src/calibre/web/feeds/recipes/recipe_iht.py

@@ -3,6 +3,7 @@
 '''
 iht.com
 '''
+import re
 
 from calibre.web.feeds.news import BasicNewsRecipe
 from calibre.ptempfile import PersistentTemporaryFile
@@ -16,7 +17,12 @@ class InternationalHeraldTribune(BasicNewsRecipe):
     max_articles_per_feed = 10
     no_stylesheets = True
 
-    remove_tags    = [dict(name='div', attrs={'class':'footer'})]
+    remove_tags    = [dict(name='div', attrs={'class':'footer'}),
+                      dict(name=['form'])]
+    preprocess_regexps = [
+            (re.compile(r'<!-- webtrends.*', re.DOTALL), 
+             lambda m:'</body></html>')
+                          ]
     extra_css      = '.headline {font-size: x-large;} \n .fact { padding-top: 10pt  }' 
 
     feeds          = [